The present invention relates to system maintenance and diagnosis, and more particularly to techniques that provide a simple and easily understandable indicator identifying the health or status of a system.
In recent years, efforts have been made to improve the manageability of large software systems, such as an Oracle database (DB) system. As part of managing the software system, information related to performance, resource utilization, and reliability are typically collected for the software system and presented to a software system administrator, such as a database administrator (DBA) for a database product.
While the information collected is useful, the volume of the information poses challenges to the software system administrator. One challenge is being able to understand, comprehend, and prioritize the information in a timely manner. Due to the volume of information and difficulty in correlating the information, many times it is not possible for an administrator to quickly determine the status of a system without having to spend a lot of time in analyzing and wading through the collected information. The problem is further aggravated when multiple systems are being managed.